Estimating global impacts from climate change

We surveyed the literature to assess the state of knowledge with regard to the (presumed) benefits or avoided damages of reducing atmospheric concentrations of greenhouse gases to progressively lower levels. The survey included only published studies addressing global impacts of climate change; studies that only addressed regional impacts were not included. The metric we used for change in climate is increase in global mean temperature (GMT). The focus of the analysis centred on determining the general shape of the damage curve, expressed as a function of GMT. Studies in sea level rise, agriculture, water resources, human health, energy, terrestrial ecosystems productivity, forestry, biodiversity, and marine ecosystems productivity were examined. In addition, we analysed several studies that aggregate results across sectors. Results are presented using metrics as reported in the surveyed studies and thus are not aggregated. We found that the relationships between GMT and impacts are not consistent across sectors. Some of the sectors exhibit increasing adverse impacts with increasing GMT, in particular coastal resources, biodiversity, and possibly marine ecosystem productivity. Some sectors are characterised by a parabolic relationship between temperature and impacts (benefits at lower GMT increases, damages at higher GMT increases), in particular, agriculture, terrestrial ecosystem productivity, and possibly forestry. The relationship between global impacts and increase in GMT for water, health, energy, and aggregate impacts appears to be uncertain. One consistent pattern is that beyond an approximate 3-4degreesC increase in GMT, all of the studies we examined, with the possible exception of forestry, show increasing adverse impacts. Thus, in total, it appears likely that there are increasing adverse impacts at higher increases in GMT. We were unable to determine the relationship between total impacts and climate change up to a 3-4degreesC increase in GMT. There are important uncertainties in the studies we surveyed that prevent us from a precise identification of 3-4degreesC as the critical temperature transition range, beyond which damages are adverse and increasing. We are confident in general however, that beyond several degrees of GMT, damages tend to be adverse and increasing. We conclude by suggesting some priorities for future research that, if undertaken, would further our understanding of how impacts are apt to vary with increases in GMT. (C) 2004 Elsevier Ltd. All rights reserved.